{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,27]],"date-time":"2026-10-27T20:26:34Z","timestamp":1793132794317,"version":"4.5.0"},"reference-count":42,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T00:00:00Z","timestamp":1780963200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T00:00:00Z","timestamp":1787270400000},"content-version":"vor","delay-in-days":73,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Early detection of Alzheimer\u2019s disease (AD) is important because of its progressive impact on cognitive function. This study presents a CNN\u2013LSTM-based framework for three-class AD classification from structural MRI, with the primary contribution being a post-hoc information-theoretic evaluation strategy rather than a new network architecture. Experiments were conducted using 827 ADNI subjects, including normal controls (NC: 340), mild cognitive impairment (MCI: 307), and AD (180). To mitigate data scarcity and improve training diversity, GAN-based augmentation was applied only to the training data, while validation and test subjects were kept separate. In addition to conventional metrics, trained models were evaluated using Renyi mutual information, Renyi divergence, and Henze\u2013Penrose divergence to quantify information preservation, representation stability, and distributional alignment. Under a subject-level evaluation protocol, the CNN\u2013LSTM model achieved 96.7% accuracy and outperformed evaluated benchmark architectures under the same protocol. The information-theoretic measures provided complementary evidence for comparing model behavior beyond accuracy, particularly regarding information retention and output-distribution alignment. Overall, the findings suggest that post-hoc information-theoretic analysis can support more transparent assessment of MRI-based AD classification models. However, external validation on independent multi-center datasets is required before clinical deployment can be considered.<\/jats:p>","DOI":"10.1038\/s41598-026-56939-y","type":"journal-article","created":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T12:55:07Z","timestamp":1781009707000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["An information-theoretic evaluation framework for CNN\u2013LSTM-based Alzheimer\u2019s disease classification from structural MRI"],"prefix":"10.1038","volume":"16","author":[{"given":"Shiva","family":"Sanati","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Elias","family":"Rahimi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ghosheh Abed","family":"Hodtani","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Saeid","family":"Eslami","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,6,9]]},"reference":[{"key":"56939_CR1","doi-asserted-by":"publisher","first-page":"2354","DOI":"10.1109\/TMI.2021.3077079","volume":"40","author":"W Zhu","year":"2021","unstructured":"Zhu, W., Sun, L., Huang, J., Han, L. & Zhang, D. Dual attention multi-instance deep learning for Alzheimer\u2019s disease diagnosis with structural MRI. IEEE Trans. Med. Imaging. 40, 2354\u20132366 (2021).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"56939_CR2","doi-asserted-by":"crossref","unstructured":"Punzi, M. et al. Atrophy of hippocampal subfields and amygdala nuclei in subjects with mild cognitive impairment progressing to Alzheimer\u2019s disease. Heliyon 10, 2563 (2024).","DOI":"10.1016\/j.heliyon.2024.e27429"},{"key":"56939_CR3","doi-asserted-by":"publisher","first-page":"177","DOI":"10.1111\/jdi.13622","volume":"13","author":"Y Ueba","year":"2022","unstructured":"Ueba, Y. et al. Voxel-based specific regional analysis system for Alzheimer\u2019s disease utility as a screening tool for unrecognized cognitive dysfunction of elderly patients in diabetes outpatient clinics: Multicenter retrospective exploratory study. J. Diabetes Invest. 13, 177\u2013184 (2022).","journal-title":"J. Diabetes Invest."},{"key":"56939_CR4","unstructured":"Payan, A. & Montana, G. Predicting Alzheimer\u2019s disease: a neuroimaging study with 3D convolutional neural networks. arXiv preprint arXiv:1502.02506 (2015)."},{"key":"56939_CR5","doi-asserted-by":"publisher","first-page":"11616","DOI":"10.1038\/s41598-025-86635-2","volume":"15","author":"MZ Hussain","year":"2025","unstructured":"Hussain, M. Z. et al. A fine-tuned convolutional neural network model for accurate Alzheimer\u2019s disease classification. Sci. Rep. 15, 11616 (2025).","journal-title":"Sci. Rep."},{"key":"56939_CR6","doi-asserted-by":"publisher","first-page":"1664","DOI":"10.1109\/TMI.2023.3344384","volume":"43","author":"K Han","year":"2023","unstructured":"Han, K., Li, G., Fang, Z. & Yang, F. Multi-template meta-information regularized network for alzheimer\u2019s disease diagnosis using structural mri. IEEE Trans. Med. Imaging. 43, 1664\u20131676 (2023).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"56939_CR7","doi-asserted-by":"publisher","first-page":"6252","DOI":"10.1109\/TNNLS.2018.2827778","volume":"29","author":"AR Heravi","year":"2018","unstructured":"Heravi, A. R. & Hodtani, G. A. A new correntropy-based conjugate gradient backpropagation algorithm for improving training in neural networks. IEEE Trans. neural networks Learn. Syst. 29, 6252\u20136263 (2018).","journal-title":"IEEE Trans. neural networks Learn. Syst."},{"key":"56939_CR8","doi-asserted-by":"crossref","unstructured":"Lombardi, G. et al. Structural magnetic resonance imaging for the early diagnosis of dementia due to Alzheimer\u2019s disease in people with mild cognitive impairment. Cochrane Database Syst. Rev. (2020).","DOI":"10.1002\/14651858.CD009628.pub2"},{"key":"56939_CR9","doi-asserted-by":"publisher","first-page":"8384","DOI":"10.1038\/s41598-020-64452-z","volume":"10","author":"S Hosseinian","year":"2020","unstructured":"Hosseinian, S. et al. A meta-analysis of gene expression data highlights synaptic dysfunction in the hippocampus of brains with Alzheimer\u2019s disease. Sci. Rep. 10, 8384 (2020).","journal-title":"Sci. Rep."},{"key":"56939_CR10","doi-asserted-by":"publisher","first-page":"3517","DOI":"10.1038\/s41467-021-23243-4","volume":"12","author":"RG Smith","year":"2021","unstructured":"Smith, R. G. et al. A meta-analysis of epigenome-wide association studies in Alzheimer\u2019s disease highlights novel differentially methylated loci across cortex. Nat. Commun. 12, 3517 (2021).","journal-title":"Nat. Commun."},{"key":"56939_CR11","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.neurobiolaging.2019.11.012","volume":"87","author":"E Lyros","year":"2020","unstructured":"Lyros, E. et al. Normal brain aging and Alzheimer\u2019s disease are associated with lower cerebral pH: an in vivo histidine 1H-MR spectroscopy study. Neurobiol. Aging. 87, 60\u201369 (2020).","journal-title":"Neurobiol. Aging"},{"key":"56939_CR12","doi-asserted-by":"crossref","unstructured":"Carmo, D., Silva, B., Yasuda, C., Rittner, L. & Lotufo, R. Hippocampus segmentation on epilepsy and Alzheimer\u2019s disease studies with multiple convolutional neural networks. Heliyon 7, 25 (2021).","DOI":"10.1016\/j.heliyon.2021.e06226"},{"key":"56939_CR13","first-page":"130","volume":"29","author":"S Sanati","year":"2019","unstructured":"Sanati, S., Saadatmand, M. & Nekooei, S. Brain structural changes caused by autism spectrum disorder based on volumetric analysis of magnetic resonance images: a review study. J. Mazandaran Univ. Med. Sci. 29, 130\u2013144 (2019).","journal-title":"J. Mazandaran Univ. Med. Sci."},{"key":"56939_CR14","first-page":"69","volume":"2","author":"VS Rad","year":"2016","unstructured":"Rad, V. S., Sanati, S., Sheikholeslami, Z. & SaadatmandTarzjan, M. A new method for medical image registration based on deformable models: application for thorax CT images. J. Iran. Association Electr. Electron. Eng. 2, 69\u201380 (2016).","journal-title":"J. Iran. Association Electr. Electron. Eng."},{"key":"56939_CR15","unstructured":"Saadatmand, M., Sanati, S. & Nekooei, S. Investigating gray-matter volume abnormalities in autism spectrum disorder. In The 5th Basic and Clinical Neuroscience Congress (2016)."},{"key":"56939_CR16","doi-asserted-by":"crossref","unstructured":"Sanati, S. & Saadatmand, M. Optimizing MR image registration for accurate brain volume measurement in children with autism spectrum disorder. In 2024 14th International Conference on Computer and Knowledge Engineering (ICCKE) 423\u2013428 (IEEE,  2024).","DOI":"10.1109\/ICCKE65377.2024.10874544"},{"key":"56939_CR17","doi-asserted-by":"crossref","unstructured":"Amin-Naji, M., Mahdavinataj, H. & Aghagolzadeh, A. Alzheimer\u2019s disease diagnosis from structural MRI using Siamese convolutional neural network. In 2019 4th International Conference on Pattern Recognition and Image Analysis (IPRIA) 75\u201379 (IEEE, 2019).","DOI":"10.1109\/PRIA.2019.8786031"},{"key":"56939_CR18","doi-asserted-by":"publisher","first-page":"109031","DOI":"10.1016\/j.patcog.2022.109031","volume":"133","author":"SQ Abbas","year":"2023","unstructured":"Abbas, S. Q., Chi, L. & Chen, Y. P. P. Transformed domain convolutional neural network for Alzheimer\u2019s disease diagnosis using structural MRI. Pattern Recogn. 133, 109031 (2023).","journal-title":"Pattern Recogn."},{"key":"56939_CR19","doi-asserted-by":"publisher","first-page":"534592","DOI":"10.3389\/fbioe.2020.534592","volume":"8","author":"G Folego","year":"2020","unstructured":"Folego, G., Weiler, M., Casseb, R. F., Pires, R. & Rocha, A. Alzheimer\u2019s disease detection through whole-brain 3D-CNN MRI. Front. Bioeng. Biotechnol. 8, 534592 (2020).","journal-title":"Front. Bioeng. Biotechnol."},{"key":"56939_CR20","doi-asserted-by":"publisher","first-page":"106079","DOI":"10.1016\/j.bspc.2024.106079","volume":"92","author":"I Ayus","year":"2024","unstructured":"Ayus, I. & Gupta, D. A novel hybrid ensemble based Alzheimer\u2019s identification system using deep learning technique. Biomed. Signal Process. Control. 92, 106079 (2024).","journal-title":"Biomed. Signal Process. Control"},{"key":"56939_CR21","doi-asserted-by":"publisher","first-page":"1140782","DOI":"10.3389\/fncom.2023.1140782","volume":"17","author":"S Sanati","year":"2023","unstructured":"Sanati, S., Rouhani, M. & Hodtani, G. A. Information-theoretic analysis of Hierarchical Temporal Memory-Spatial Pooler algorithm with a new upper bound for the standard information bottleneck method. Front. Comput. Neurosci. 17, 1140782 (2023).","journal-title":"Front. Comput. Neurosci."},{"key":"56939_CR22","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1007\/s11063-024-11546-8","volume":"56","author":"S Sanati","year":"2024","unstructured":"Sanati, S., Rouhani, M. & Hodtani, G. A. Performance comparison of different HTM-spatial pooler algorithms based on information-theoretic measures. Neural Process. Lett. 56, 44 (2024).","journal-title":"Neural Process. Lett."},{"key":"56939_CR23","doi-asserted-by":"publisher","first-page":"109228","DOI":"10.1016\/j.patcog.2022.109228","volume":"136","author":"F Yuan","year":"2023","unstructured":"Yuan, F., Zhang, Z. & Fang, Z. An effective CNN and Transformer complementary network for medical image segmentation. Pattern Recogn. 136, 109228 (2023).","journal-title":"Pattern Recogn."},{"key":"56939_CR24","doi-asserted-by":"publisher","first-page":"102","DOI":"10.1016\/j.zemedi.2018.11.002","volume":"29","author":"AS Lundervold","year":"2019","unstructured":"Lundervold, A. S. & Lundervold, A. An overview of deep learning in medical imaging focusing on MRI. Z. fuer medizinische Physik. 29, 102\u2013127 (2019).","journal-title":"Z. fuer medizinische Physik"},{"key":"56939_CR25","doi-asserted-by":"crossref","unstructured":"Manzak, D., \u00c7etinel, G. & Manzak, A. Automated classification of Alzheimer\u2019s disease using deep neural network (DNN) by random forest feature elimination. In 2019 14th International Conference on Computer Science & Education (ICCSE) 1050\u20131053 (IEEE,  2019).","DOI":"10.1109\/ICCSE.2019.8845325"},{"key":"56939_CR26","doi-asserted-by":"publisher","first-page":"206","DOI":"10.1109\/LSP.2020.2964161","volume":"27","author":"JY Choi","year":"2020","unstructured":"Choi, J. Y. & Lee, B. Combining of multiple deep networks via ensemble generalization loss, based on MRI images, for Alzheimer\u2019s disease classification. IEEE. Signal. Process. Lett. 27, 206\u2013210 (2020).","journal-title":"IEEE. Signal. Process. Lett."},{"key":"56939_CR27","unstructured":"Sarraf, S. & Tofighi, G. Classification of Alzheimer\u2019s disease structural MRI data by deep learning convolutional neural networks. arXiv preprint arXiv:1607.06583 (2016)."},{"key":"56939_CR28","doi-asserted-by":"publisher","first-page":"9087","DOI":"10.1038\/s41598-023-36311-0","volume":"13","author":"Z Ullah","year":"2023","unstructured":"Ullah, Z., Usman, M., Latif, S., Khan, A. & Gwak, J. SSMD-UNet: semi-supervised multi-task decoders network for diabetic retinopathy segmentation. Sci. Rep. 13, 9087 (2023).","journal-title":"Sci. Rep."},{"key":"56939_CR29","doi-asserted-by":"publisher","first-page":"3728","DOI":"10.3390\/math13223728","volume":"13","author":"Z Ullah","year":"2025","unstructured":"Ullah, Z., Hong, M., Mahmood, T. & Kim, J. Systematic integration of attention modules into CNNs for accurate and generalizable medical image classification. Mathematics 13, 3728 (2025).","journal-title":"Mathematics"},{"key":"56939_CR30","doi-asserted-by":"publisher","first-page":"2787","DOI":"10.3390\/math13172787","volume":"13","author":"Z Ullah","year":"2025","unstructured":"Ullah, Z. & Kim, J. Hierarchical deep feature fusion and ensemble learning for enhanced brain tumor MRI classification. Mathematics 13, 2787 (2025).","journal-title":"Mathematics"},{"key":"56939_CR31","doi-asserted-by":"publisher","first-page":"e3335","DOI":"10.7717\/peerj-cs.3335","volume":"11","author":"W Abdelbaki","year":"2025","unstructured":"Abdelbaki, W. et al. Efficient hybrid CNN-transformer model for accurate blood cancer detection. PeerJ Comput. Sci. 11, e3335 (2025).","journal-title":"PeerJ Comput. Sci."},{"key":"56939_CR32","doi-asserted-by":"publisher","first-page":"111058","DOI":"10.1016\/j.engappai.2025.111058","volume":"155","author":"SY Lu","year":"2025","unstructured":"Lu, S. Y., Zhang, Y. D. & Yao, Y. D. A regularized transformer with adaptive token fusion for Alzheimer\u2019s disease diagnosis in brain magnetic resonance images. Eng. Appl. Artif. Intell. 155, 111058 (2025).","journal-title":"Eng. Appl. Artif. Intell."},{"key":"56939_CR33","doi-asserted-by":"publisher","first-page":"120764","DOI":"10.1016\/j.measurement.2026.120764","volume":"269","author":"SY Lu","year":"2026","unstructured":"Lu, S. Y., Zhang, Y. D. & Yao, Y. D. Integrating segment anything model with hierarchical enhanced linear attention for Alzheimer\u2019s disease detection. Measurement 269, 120764 (2026).","journal-title":"Measurement"},{"key":"56939_CR34","doi-asserted-by":"publisher","first-page":"111575","DOI":"10.1016\/j.engappai.2025.111575","volume":"158","author":"SY Lu","year":"2025","unstructured":"Lu, S. Y., Zhu, Z., Zhang, Y. D. & Yao, Y. D. Tuberculosis and pneumonia diagnosis in chest X-rays by large adaptive filter and aligning normalized network with report-guided multi-level alignment. Eng. Appl. Artif. Intell. 158, 111575 (2025).","journal-title":"Eng. Appl. Artif. Intell."},{"key":"56939_CR35","doi-asserted-by":"publisher","first-page":"e092170","DOI":"10.1002\/alz.092170","volume":"20","author":"MM Bouwman","year":"2024","unstructured":"Bouwman, M. M., Frigerio, I., Reijner, N., van de Berg, W. D. & Jonkman, L. E. Synaptic density in the hippocampal and parahippocampal subregions and its association with the severity of axonal damage and cognitive decline in Alzheimer\u2019s disease. Alzheimer\u2019s Dement. 20, e092170 (2024).","journal-title":"Alzheimer\u2019s Dement."},{"key":"56939_CR36","doi-asserted-by":"publisher","first-page":"37","DOI":"10.1007\/s00401-024-02789-9","volume":"148","author":"AE Denning","year":"2024","unstructured":"Denning, A. E. et al. Association of quantitative histopathology measurements with antemortem medial temporal lobe cortical thickness in the Alzheimer\u2019s disease continuum. Acta Neuropathol. 148, 37 (2024).","journal-title":"Acta Neuropathol."},{"key":"56939_CR37","doi-asserted-by":"publisher","first-page":"2600","DOI":"10.1109\/TNSRE.2022.3202713","volume":"30","author":"H Huang","year":"2022","unstructured":"Huang, H. et al. Deep spatio-temporal attention-based recurrent network from dynamic adaptive functional connectivity for MCI identification. IEEE Trans. Neural Syst. Rehabil. Eng. 30, 2600\u20132612 (2022).","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"56939_CR38","doi-asserted-by":"publisher","first-page":"41","DOI":"10.17977\/um018v2i12019p41-46","volume":"2","author":"IKM Jais","year":"2019","unstructured":"Jais, I. K. M., Ismail, A. R. & Nisa, S. Q. Adam optimization algorithm for wide and deep neural network. Knowl. Eng. Data Sci. 2, 41\u201346 (2019).","journal-title":"Knowl. Eng. Data Sci."},{"key":"56939_CR39","doi-asserted-by":"publisher","first-page":"40","DOI":"10.3758\/BF03213026","volume":"9","author":"JT Townsend","year":"1971","unstructured":"Townsend, J. T. Theoretical analysis of an alphabetic confusion matrix. Percept. Psychophys. 9, 40\u201350 (1971).","journal-title":"Percept. Psychophys."},{"key":"56939_CR40","first-page":"538","volume":"532","author":"P Refaeilzadeh","year":"2009","unstructured":"Refaeilzadeh, P., Tang, L. & Liu, H. Cross-validation. Encyclopedia Database Syst. 532, 538 (2009).","journal-title":"Encyclopedia Database Syst."},{"key":"56939_CR41","doi-asserted-by":"publisher","first-page":"174","DOI":"10.1007\/s12559-025-10521-2","volume":"17","author":"MJ Abbas","year":"2025","unstructured":"Abbas, M. J. et al. XRDNet: a novel explainable residual dense fusion network for alzheimer\u2019s disease recognition from MRI images. Cogn. Comput. 17, 174 (2025).","journal-title":"Cogn. Comput."},{"key":"56939_CR42","doi-asserted-by":"publisher","first-page":"e70223","DOI":"10.1111\/exsy.70223","volume":"43","author":"MJ Abbas","year":"2026","unstructured":"Abbas, M. J. et al. A Novel Multistage Attention-Enhanced Mixture of Experts Model for Alzheimer\u2019s Disease Diagnosis. Expert Syst. 43, e70223 (2026).","journal-title":"Expert Syst."}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-56939-y","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-56939-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-56939-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T22:57:12Z","timestamp":1787266632000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-56939-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,9]]},"references-count":42,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["56939"],"URL":"https:\/\/doi.org\/10.1038\/s41598-026-56939-y","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,9]]},"assertion":[{"value":"9 February 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 June 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"9 June 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}},{"value":"This study used de-identified, previously collected data from the Alzheimer\u2019s Disease Neuroimaging Initiative (ADNI) database. The ADNI study protocols were approved by the Institutional Review Board of the University of Southern California (USC IRB) and by the institutional review boards\/ethics committees of the participating ADNI clinical sites. Written informed consent was obtained from all participants and\/or their legally authorized representatives by the ADNI investigators at the participating sites prior to enrollment. The present analysis did not involve new participant recruitment, new data collection, direct participant contact, intervention, access to identifiable personal information, or redistribution of participant-level data. No additional participant recruitment or informed-consent procedure was undertaken by the authors for this secondary analysis. All methods were performed in accordance with the relevant guidelines and regulations, including the Declaration of Helsinki, Good Clinical Practice guidelines, applicable institutional regulations, and ADNI data-use policies.","order":2,"name":"Ethics","label":"Ethics approval","group":{"name":"EthicsHeading","label":"Ethics"}}],"article-number":"26240"}}